4

Is there any attempt to gauge how quickly a scientific field overturns its own claims/findings; ie an evaluation of if the fields’ findings have what degree of longevity?

It could compare fields as diverse as physics, psychology, and literary theory.

3
  • 1
    What constitutes a finding? What constitutes an overturning? It’s not as though there were a Great Clearinghouse of Astronomy—or of any other science—and a bipartition {Yes, No} of the set of all propositions into the enshrined and the forbidden. Sep 11 at 19:57
  • 1
    There is statistics of retractions, see Self-correction in science by Kühberger et al., but that concerns claims made at the level of individual researchers or groups. As for claims that reach broad acceptance at some point, whether they are "overturned" is controversial. Oxidation "overturned" phlogiston, say, or relativity "overturned" ether is one way to put it, but newer theories typically make predictions consistent with older theories within the range of applicability of the latter. It is unclear what to track there statistically.
    – Conifold
    Sep 12 at 10:39
  • 1
    I suggest adding "computer science" to your list of fields to compare. Being much much younger than other fields, it's bound to be a lot less "stable".
    – Stef
    Sep 12 at 15:27

2 Answers 2

3

Paradigm shifts are Kuhn's model of such change, as JD has referenced. His picture of a power struggle and a kind of 'democracy' of most popular ideas or frameworks taking over, is highly disputed. But the term has entered the language, and others have developed more evidence-based updates to it. See eg SEP Scientific Revolutions - Larger Formations and Historical A Prioris: The Germanic and French Traditions.

These are really more narrative rather than quantative though - I'm not sure how much could be gained from the latter. I remember reading about applied chaos theory and complex systems, which might be a case where it would be useful - these can describe phase transitions like from laminar to turbulent flow, which better mach real changes (see punctuated equilbria in evolution too). Patent applications is often used as a proxy for technological development. And you might use papers that meet certain criteria like high and persistent citatiins as a proxy for development in a discipline. You need to be clear why you want to know, and what kind of answer you'd find satisfying. Are you just looking for dynamism? Or fields that are best established and change least? What conclusions do you expect or hope to draw?

The impact of technology is going to be important. Lots of business culture updates are being driven by the world if coding, like say the shift to agile development. Any field very affected by computing and data analysis will be changing rapidly. Schumpeter's model of successive technological cycles is an interesting approach. See Schumpeter cycles. There's also the similar Kondratiev waves. If we look at the history of science, innovations in lense-making (telescopes, microscopes, & optics), and the air pump, had epochal impacts in accelerating research.

3

Yes. As far as my limited knowledge permits, there are two domains where you might search for this related to the idea that comes from the Kuhnian attack on the objectivity and linearity of scientific practice. The first is history of science, and the second is the sociology of science. Neither of those disciplines, strictly speaking are philosophy sub-disciplines if one observes the political hierarchy, however, both topics are very indebted to the philosophy of science, and I would start with the PhilPapers search on the 'history of science' for academics who have invested themselves in developing expertise for evaluation of claims around the terminology of 'normal and paradigmatic science'.

What constitutes a finding? What constitutes an overturning? It’s not as though there were a Great Clearinghouse of Astronomy—or of any other science—and a bipartition {Yes, No} of the set of all propositions into the enshrined and the forbidden. – Paul Tanenbaum

Obviously the operationalization of definition would be something that would emerge as a preliminary issue, and if you manage to find some academic papers or books that do so, I'd sure be interested in some quick pointers. Come back and post your own answer! :D

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .